Bounded-parameter Markov decision processes.
Robert GivanSonia M. LeachThomas L. DeanPublished in: Artif. Intell. (2000)
Keyphrases
- markov decision processes
- state space
- finite state
- optimal policy
- reinforcement learning
- dynamic programming
- reachability analysis
- decision theoretic planning
- average cost
- transition matrices
- finite horizon
- policy iteration
- reinforcement learning algorithms
- action space
- decision processes
- partially observable
- planning under uncertainty
- factored mdps
- markov decision process
- reward function
- average reward
- action sets
- risk sensitive
- state abstraction
- multistage
- linear programming
- probability distribution